This proposal describes a major new initiative to develop synthetic antibodies by a "mono-molecular" imprinting process. Molecular imprinted polymers (MIPs) are well-established materials widely acknowledged for their extraordinary potential to impact biotechnology and biomedicine. These materials are formed by carrying out a polymerization reaction in the presence of a template. However, several limitations not yet overcome, have prevented MIPs from achieving commercial applicability. Principal among these limitations is binding site heterogeneity, slow mass transfer, insolubility, difficulty in quantitatively removing the template, and an inability of most MEN to function in water. The proposed effort will develop a "molding" protocol in which a single polymeric structure, covalently connected to a template, will be extensively cross-linked. Covalent cleavage of the template will leave a nanostructure containing a single binding site that is both shape-- selective and complementary in its functional group array to the template molecule (antigen). The templates to be investigated in this early phase of the project include carbohydrates, amino acids, peptides, and drug molecules. Several approaches are borrowed directly from polymer imprinting for comparison purposes. The initial polymers to be studied will be dendritic in structure. A major advantage of this approach is that although multiple compounds will be produced (polyclonal approach) one might separate the most efficacious compounds. Furthermore, these "synthetic antibodies" are expected to show rapid binding kinetics, be easily modified chemically (e.g., solubility in a range of media, attachment to surfaces), and to have readily removable templates. Infinite modifications in the polymer structure, solvent, type and degree of cross-linking makes it very likely that this approach will be successful, and that the binding strength and selectivity can be tuned. Applications in medical diagnostics are the most likely early pay-off for this technology.